Clinical Actionability Research Articles

P17.06.B TARGETED THERAPY MATCHED TO GENOMIC ALTERATIONS (BRAFV600E, NTRK, FGFR, ROS1, PIK3CA, PTEN) IN PATIENTS WITH RECURRENT IDH WILDTYPE GLIOBLASTOMA: A REAL-LIFE COHORT ANALYSIS FROM VENETO INSTITUTE OF ONCOLOGY, PADUA (ITALY)

Abstract BACKGROUND The use of next-generation sequencing (NGS) allows the identification of subgroups of patients that may respond to personalized targeted therapies (TT). Despite the acquired deep knowledge about biology, there are only a few data available regarding TT efficacy and feasibility for patients with glioblastoma (GBM). MATERIAL AND METHODS We retrospectively analyzed a cohort of patients with recurrent GBM treated at Veneto Institute of Oncology, Padua (Italy) with TT based on their individual NGS profile. NGS was obtained with FoundationOne®CDx on formalin-fixed paraffin-embedded (FFPE) tissue samples. We identified 6 druggable genomic alterations, classified according to ESCAT (ESMO Scale for Clinical Actionability of molecular Targets): BRAFV600E mutation (IB), NTRK1-2-3 fusions (IC), FGFR1-2-3 alterations (IIB), ROS1 fusions (IIIA), PIK3CA mutations (IIIA) and PTEN loss/mutations (IIIA). TT was given as agnostic approval, part of compassionate use programs or in clinical trials. Response assessment was based on RANO (Response Assessment in Neuro-Oncology) criteria. RESULTS Between March 2020 and December 2022, 34 patients with recurrent glioblastoma received TT (19 males, 15 females, median age of 54 years at time of TT initiation). ECOG PS was 0-1 in 29 pts. All patients received radiotherapy and temozolomide as first-line therapy. The median line of therapy with TT was 3 (range 2-7). TT was dabrafenib/trametinib (9 pts), larotrectinib (2 pts), erdafitinib (4 pts), entrectinib (1 pt), alpelisib (6 pts), ipatasertib+/-atezolizumab (12 pts). At data cut-off (March 2023), 19 patients had died. In the entire cohort, median overall survival and progression-free survival (PFS) after starting TT was 8.72 and 2.14 months, respectively. The dabrafenib/trametinib subgroup had the longest median PFS (5.23 months), a disease control rate (DCR) of 77%, an objective response rate of 22%, and a median duration of response of 22.5 months. 7/9 pt had died and 2 pts are continuing dabrafenib/trametinib. The pt with ROS1-GOCP fusion maintained a complete response for 12 months with entrectinib. Among the others, no complete/partial responses were detected. DCR due to stable disease was 50% in larotrectinib and erdafitinib, 8.3% in ipatasertib+/-atezolizumab, 0% in alpelisib subgroup. No toxicities were reported in dabrafenib/trametinib subgroup. Among all patients, no grade 4 drug-related adverse events were observed and in any case TT was interrupted for toxicity. CONCLUSION Our results confirm the activity of dabrafenib/trametinib in BRAFV600E mutant glioblastoma pts and might suggest deeper explorations in targeting ROS1 and FGFR. Further correlation among patients' outcome, molecular characteristics and TT timing are still under investigation.

Liquid biopsy comprehensive genomic profiling of lung cancer in the Italian population: A real-world experience

ObjectivesLiquid biopsy with next-generation sequencing (NGS) has emerged as a promising tool for tumor mutation profiling. In this study, we describe the genomic profile of Italian lung cancer patients tested with blood-based comprehensive genomic profiling (CGP) to assess the genomic landscape complexity and its impact on enhancing treatment options for patients. Materials and methodsBetween January 2021 and December 2021, a total of 229 lung cancer patients were profiled by FoundationOne®Liquid CDx (F1LCDx®) assay on circulating tumor DNA (ctDNA). F1LCDx® reports alterations across 324 cancer-related genes and genomic signatures, including tumor fraction (TF) and blood-based tumor mutational burden (bTMB). Detected variants were classified according to the ESMO Scale of Clinical Actionability for molecular Targets (ESCAT). Results90.4% of patients had at least one detectable alteration in plasma. The most frequently mutated genes were TP53 (47.6%), DNMT3A (33.2%), EGFR (20.1%), and KRAS (15.7%). Elevated TF was detected in 18.3% of patients, suggesting high reliability of test results. According to the ESCAT classification, potentially actionable alterations (Tier I-II) were identified in 27.1% of samples. An additional 5.2% harbored an alteration for which an approved drug is available in other cancer types (Tier III). Furthermore, 13.1% of tumors exhibited high bTMB, which may predict response to immunotherapy. Overall, 156 (68.1%) patients were eligible for enrolment in clinical trials. ConclusionLiquid biopsy NGS is a viable and valuable approach to guide personalized therapy. The use of blood-based CGP may help identify a larger number of actionable mutations and increase chances of enrolment in clinical trials.

Rate of Pathogenic Germline Variants in Patients With Lung Cancer

PURPOSE Germline genetic testing (GGT) is now recommended for all patients diagnosed with ovarian or pancreatic cancer and for a large proportion of patients based solely on a diagnosis of colorectal or breast cancer. However, GGT is not yet recommended for all patients diagnosed with lung cancer (LC), primarily because of a lack of evidence that supports a significant frequency of identifying pathogenic germline variants (PGVs) in these patients. This study characterizes GGT results in a cohort of patients with LC. METHODS We reviewed deidentified data for 7,788 patients with GGT (2015-2022). PGV frequencies were compared to a control cohort of unaffected individuals. GGT results were stratified by genomic ancestry, history of cancer, and PGV clinical actionability per current guidelines. RESULTS Of all patients with LC, 14.9% (1,161/7,788) had PGVs. The rate was similar when restricted to patients with no cancer family history (FH) or personal history (PH) of other cancers (14.3%). PGVs were significantly enriched in BRCA2, ATM, CHEK2, BRCA1, and mismatch repair genes compared with controls. Patients of European (EUR) genomic ancestry had the highest PGV rate (18%) and variants of uncertain significance were significantly higher in patients of non-EUR genomic ancestry. Of the PGVs identified, 61.3% were in DNA damage repair (DDR) genes and 95% were clinically actionable. CONCLUSION This retrospective study shows a LC diagnosis identifies patients with a significant likelihood of having a cancer-predisposing PGV across genomic ancestries. Enrichment of PGVs in DDR genes suggests that these PGVs may contribute to LC cancer predisposition. The frequency of PGVs among patients with LC did not differ significantly according to FH or PH of other cancers.

Whole Exome Analysis to Select Targeted Therapies for Patients with Metastatic Breast Cancer - A Feasibility Study.

The purpose of this feasibility study was to select targeted therapies according to "ESMO Scale for Clinical Actionability of molecular Targets (ESCAT)". Data interpretation was further supported by a browser-based Treatment Decision Support platform (MH Guide, Molecular Health, Heidelberg, Germany). We applied next generation sequencing based whole exome sequencing of tumor tissue and peripheral blood of patients with metastatic breast cancer (n=44) to detect somatic as well as germline mutations. In 32metastatic breast cancer patients, data interpretation was feasible. We identified 25 genomic alterations with ESCAT Level of Evidence I or II in 18/32 metastatic breast cancer patients, which were available for evaluation: three copy number gains in HER2 , two g BRCA1 , two g BRCA2 , six PIK3CA, one ESR1 , three PTEN , one AKT1 and two HER2 mutations. In addition, five samples displayed Microsatellite instability high-H. Resulting treatment options were discussed in a tumor board and could be recommended in a small but relevant proportion of patients with metastatic breast cancer (7/18). Thus, this study is a valuable preliminary work for the establishment of a molecular tumor board within the German initiative "Center for Personalized Medicine" which aims to shorten time for analyses and optimize selection of targeted therapies.

The Landscape of Alterations from 1407 Ultra-Rare Sarcomas from the AACR GENIE Database: Clinical Implications.

Ultra-rare sarcomas (URS) comprise a group of orphan diseases with an incidence of ≤1/1,000,000 people per year. We aimed to assess clinically actionable genomic alterations in URS. Data were extracted from the GENIE database using cBioPortal. OncoKB was used to assess for clinical actionability of mutations. Tumor mutational burden (TMB) was inferred from clinical sequencing data. Soft tissue (ST) URS made up 23.5% of ST sarcoma cases, and bone URS made up 16.5% of bone sarcoma cases. The most commonly mutated gene in all four groups was TP53. The most common fusions involved EWSR1. The most common copy-number variations included deletions of CDKN2A and CDKN2B and amplifications of MDM2 and CDK4. TMB was generally low across all four categories of sarcoma, though there was considerable heterogeneity, with 3.8% of ST URS and 0.55% of bone URS having high TMB. We find Level 1 alterations (FDA-recognized biomarker predictive of response to an FDA-approved drug) in 10.0% of ST URS compared with 7.1% of ST non-URS, 1.1% of bone URS, and 4.5% of bone non-URS. Level 1-3 alterations (also include alterations for which there are standard-of-care drugs or clinical evidence supporting a drug) were seen in 27.8% of ST URS, 25.2% of ST non-URS, 20.9% of bone URS, and 17.4% of bone non-URS. Clinically actionable genomic alterations are seen in a substantial fraction of URS. Clinical sequencing in advanced URS has the potential to guide the treatment of a significant portion of patients with URS.

Assessment of clinically actionable pharmacogenetic markers to stratify anti-seizure medications.

Epilepsy treatment is challenging due to heterogeneous syndromes, different seizure types and higher inter-individual variability. Identification of genetic variants predicting drug efficacy, tolerability and risk of adverse-effects for anti-seizure medications (ASMs) is essential. Here, we assessed the clinical actionability of known genetic variants, based on their functional and clinical significance and estimated their diagnostic predictability. We performed a systematic PubMed search to identify articles with pharmacogenomic (PGx) information for forty known ASMs. Functional annotation of the identified genetic variants was performed using different in silico tools, and their clinical significance was assessed using the American College of Medical Genetics (ACMG) guidelines for variant pathogenicity, level of evidence (LOE) from PharmGKB and the United States-Food and drug administration (US- FDA) drug labelling with PGx information. Diagnostic predictability of the replicated genetic variants was evaluated by calculating their accuracy. A total of 270 articles were retrieved with PGx evidence associated with 19 ASMs including 178 variants across 93 genes, classifying 26 genetic variants as benign/ likely benign, fourteen as drug response markers and three as risk factors for drug response. Only seventeen of these were replicated, with accuracy (up to 95%) in predicting PGx outcomes specific to six ASMs. Eight out of seventeen variants have FDA-approved PGx drug labelling for clinical implementation. Therefore, the remaining nine variants promise for potential clinical actionability and can be improvised with additional experimental evidence for clinical utility.

The Value of Pharmacogenomics for White and Indigenous Americans after Kidney Transplantation.

There is a paucity of evidence to inform the value of pharmacogenomic (PGx) results in patients after kidney transplant and how these results differ between Indigenous Americans and Whites. This study aims to identify the frequency of recommended medication changes based on PGx results and compare the pharmacogenomic (PGx) results and patients' perceptions of the findings between a cohort of Indigenous American and White kidney transplant recipients. Thirty-one Indigenous Americans and fifty White kidney transplant recipients were studied prospectively. Genetic variants were identified using the OneOme RightMed PGx test of 27 genes. PGx pharmacist generated a report of the genetic variation and recommended changes. Pre- and post-qualitative patient surveys were obtained. White and Indigenous American subjects had a similar mean number of medications at the time of PGx testing (mean 13 (SD 4.5)). In the entire cohort, 53% received beta blockers, 30% received antidepressants, 16% anticoagulation, 47% pain medication, and 25% statin therapy. Drug-gene interactions that warranted a clinical action were present in 21.5% of patients. In 12.7%, monitoring was recommended. Compared to the Whites, the Indigenous American patients had more normal CYP2C19 (p = 0.012) and CYP2D6 (p = 0.012) activities. The Indigenous American patients had more normal CYP4F2 (p = 0.004) and lower VKORC (p = 0.041) activities, phenotypes for warfarin drug dosing, and efficacy compared to the Whites. SLC6A4, which affects antidepressant metabolism, showed statistical differences between the two cohorts (p = 0.017); specifically, SLC6A4 had reduced expression in 45% of the Indigenous American patients compared to 20% of the White patients. There was no significant difference in patient perception before and after PGx. Kidney transplant recipients had several drug-gene interactions that were clinically actionable; over one-third of patients were likely to benefit from changes in medications or drug doses based on the PGx results. The Indigenous American patients differed in the expression of drug-metabolizing enzymes and drug transporters from the White patients.

SYST-20 TARGETED THERAPY MATCHED TO GENOMIC ALTERATIONS (BRAFV600E, NTRK, FGFR, ROS1, PIK3CA, PTEN) IN PATIENTS WITH RECURRENT IDH WILDTYPE GLIOBLASTOMA: A REAL-LIFE COHORT ANALYSIS FROM VENETO INSTITUTE OF ONCOLOGY, PADUA (ITALY)

Abstract We retrospectively analyzed 34 patients (pts) [19 males, 15 females, median age 54] with recurrent IDH wildtype glioblastoma treated with target therapy (TT) based on their next-generation sequencing (NGS) profile, between March2020 and December2022 at Veneto Institute of Oncology, Padua (Italy). ECOG PS was 0-1 in 29 pts. All pts received previous radiotherapy and temozolomide. NGS was obtained with FoundationOne®CDx on formalin-fixed paraffin-embedded samples. We identified 6 druggable genomic alterations, classified according to ESCAT (ESMO Scale for Clinical Actionability of molecular Targets): BRAFV600E mutation (IB), NTRK1-2-3 fusions (IC), FGFR1-2-3 alterations (IIB), ROS1 fusions (IIIA), PIK3CA mutations (IIIA) and PTEN loss/mutations (IIIA). The median line of therapy with TT was 3 (range 2-7). TT was dabrafenib/trametinib (9 pts), larotrectinib (2 pts), erdafitinib (4 pts), entrectinib (1 pt), alpelisib (6 pts), ipatasertib+/-atezolizumab (12 pts). At data cut-off (March2023), 19 patients had died. In the entire cohort, median overall survival and progression-free survival (PFS) after starting TT was 8.72 and 2.14 months, respectively. The dabrafenib/trametinib subgroup had the longest median PFS (5.23 months), a disease control rate (DCR) of 77%, an objective response rate of 22%, and a median duration of response of 22.5 months. 7/9 pt had died and 2 pts are continuing dabrafenib/trametinib. The pt with ROS1-GOCP fusion maintained a complete response for 12 months with entrectinib. Among the others, no complete/partial responses were detected. DCR due to stable disease was 50% in larotrectinib and erdafitinib subgroups and 8.3% in ipatasertib+/-atezolizumab subgroup. No toxicities were reported in dabrafenib/trametinib subgroup. Among all patients, no grade 4 drug-related adverse events were observed and in any case TT was interrupted for toxicity. Our results confirm the activity of dabrafenib/trametinib in BRAFV600E mutant glioblastoma pts and might suggest further explorations in targeting ROS1 and FGFR.

Molecular Tumor Board as a Clinical Tool for Converting Molecular Data Into Real-World Patient Care.

The investigation of multiple molecular targets with next-generation sequencing (NGS) has entered clinical practice in oncology, yielding to a paradigm shift from the histology-centric approach to the mutational model for personalized treatment. Accordingly, most of the drugs recently approved in oncology are coupled to specific biomarkers. One potential tool for implementing the mutational model of precision oncology in daily practice is represented by the Molecular Tumor Board (MTB), a multidisciplinary team whereby molecular pathologists, biologists, bioinformaticians, geneticists, medical oncologists, and pharmacists cooperate to generate, interpret, and match molecular data with personalized treatments. Since May 2020, the institutional MTB set at Fondazione IRCCS Istituto Nazionale Tumori of Milan met weekly via teleconference to discuss molecular data and potential therapeutic options for patients with advanced/metastatic solid tumors. Up to October 2021, among 1,996 patients evaluated, we identified >10,000 variants, 43.2% of which were functionally relevant (pathogenic or likely pathogenic). On the basis of functionally relevant variants, 711 patients (35.6%) were potentially eligible to targeted therapy according to European Society of Medical Oncology Scale for Clinical Actionability of Molecular Targets tiers, and 9.4% received a personalized treatment. Overall, larger NGS panels (containing >50 genes) significantly outperformed small panels (up to 50 genes) in detecting actionable gene targets across different tumor types. Our real-world data provide evidence that MTB is a valuable tool for matching NGS data with targeted treatments, eventually implementing precision oncology in clinical practice.

Age-Based Genomic Screening during Childhood: Ethical and Practical Considerations in Public Health Genomics Implementation.

Genomic sequencing offers an unprecedented opportunity to detect inherited variants that are implicated in rare Mendelian disorders, yet there are many challenges to overcome before this technology can routinely be applied in the healthy population. The age-based genomic screening (ABGS) approach is a novel alternative to genome-scale sequencing at birth that aims to provide highly actionable genetic information to parents over the course of their child's routine health care. ABGS utilizes an established metric to identify conditions with high clinical actionability and incorporates information about the age of onset and age of intervention to determine the optimal time to screen for any given condition. Ongoing partnerships with parents and providers are instrumental to the co-creation of educational resources and strategies to address potential implementation barriers. Implementation science frameworks and informative empirical data are used to evaluate strategies to establish this unique clinical application of targeted genomic sequencing. Ultimately, a pilot project conducted in primary care pediatrics clinics will assess patient and implementation outcomes, parent and provider perspectives, and the feasibility of ABGS. A validated, stakeholder-informed, and practical ABGS program will include hundreds of conditions that are actionable during infancy and childhood, setting the stage for a longitudinal implementation that can assess clinical and health economic outcomes.

Actionable molecular alterations in newly diagnosed and recurrent IDH1/2 wild-type glioblastoma patients and therapeutic implications: a large mono-institutional experience using extensive next-generation sequencing analysis

BackgroundNext-generation sequencing (NGS) panels enable the identification of alterations in cancer-related genes. This may guide a molecularly targeted strategy for the treatment of glioblastoma (GBM). Material and methodsWe retrospectively analysed data obtained using FoundationOne®CDx in a large cohort of IDH1/2 wild-type GBM. We aimed to 1) identify potentially actionable molecular alterations at diagnosis and/or recurrence based on ESMO Scale for Clinical Actionability of Molecular Targets (ESCAT) defined categories of targetability, 2) understand the clinical implications of NGS in terms of access to and activity of targeted therapies. ResultsIn 442 samples, an NGS profile was available in 98.2%. The median time from diagnosis to NGS profiling was 7.4 months (interquartile range (IQR): 3.4–13.2). Although about half of the patients had at least one actionable molecular alteration, only 3.4% of them were classified as ESCAT IB–IC and 6.7% as ESCAT IIB. Only 36 patients (10.5%) received personalised treatment in clinical trials or as off-label/compassionate use from second-line (median line 3). Most patients did not receive targeted therapy due to clinical deterioration/death (49.6%). Patients treated with dabrafenib/trametinib (9 patients) had the highest disease control rate of 77% and an objective response rate of 22%, with a median progression-free survival (PFS) of 5.2 months. No complete/partial responses were seen with the other regimens. 4/9 (44.4%) patients on anti-BRAF/anti-MEK, 2/4 patients (50%) on erdafitinib and 1/1 patient on capmatinib had a PFS ratio > 1.3. One recurrent GBM patient with ROS1-GOCP fusion maintained a complete response for 11.3 months on entrectinib. ConclusionsOur study demonstrated the feasibility of NGS in GBM samples. As the number of clinically relevant targets was limited and only a small group of GBM patients were treated with targeted therapy, NGS testing should be performed in the context of clinical trials. Our results support the activity of anti-BRAF/anti-MEK, while for the other agents prospective study results are needed to draw solid conclusions.

Genomically matched therapy in refractory colorectal cancer according to ESMO Scale for Clinical Actionability of Molecular Targets: experience of a comprehensive cancer center network.

Efficiency of expanded genomic profiling (EGP) programmes in terms of final inclusion of patients in genomically matched therapies is still unknown. Fit patients with advanced and refractory colorectal cancer (CRC) were selected for an EGP programme. Next-generation sequencing (NGS) analysis from formalin-fixed paraffin-embedded tumour samples was performed. The purpose was to describe the prevalence of genomic alterations defined by the ESMO Scale for Clinical Actionability of Molecular Targets (ESCAT), as well as the percentage of patients finally included in genomically guided clinical trials. In total, 187 patients were recruited. Mutational profile was obtained in 177 patients (10 patients were failure due to insufficient tumour sample), copy number alterations in 41 patients and fusions in 31 patients. ESCAT-defined alterations were detected in 28.8% of the intention-to-analyse population. BRAF V600E was clustered in ESCAT I, with a prevalence of 3.7%, KRAS G12C and ERBB2 amplification were clustered in ESCAT II, whose prevalence was 4.2% and 1.6%, respectively. Most alterations were classified in ESCAT III (mutations in ERBB2, PIK3CA or FGFR genes and MET amplification) and IV (mutations in BRAF non-V600E, ERBB3, FBXW7, NOTCH, RNF43), with a single prevalence under 5%, except for PIK3CA mutation (9%).The final rate of inclusion into genomically guided clinical trials was 2.7%, including therapies targeting BRAF V600E or RNF43 mutations in two patients each, and ERBB2 mutation in one patient. In conclusion, EGP programmes in patients with advanced CRC are feasible and identify a subset of patients with potentially druggable genomic alterations. However, further efforts must be made to increase the rate of patients treated with genomically guided therapies.

Quantifying the growth of clinical actionability in precision oncology using OncoKB.

1577 Background: OncoKB ( www.oncokb.org ), Memorial Sloan Kettering's (MSK) precision oncology knowledge base, contains curated information about the oncogenic effect and therapeutic implications of somatic alterations in cancer and is the first cancer variant database to be partially recognized by the US-FDA. Individual mutations and structural alterations in OncoKB are assigned a level of evidence based on whether the alteration is a predictive biomarker of response or resistance to a genomically matched standard care or investigational drug in a specific cancer subtype. As of February 2023, OncoKB includes annotations for >6200 alterations in >700 cancer-associated genes, including 50 level 1 or 2 standard care genes (specified in the FDA drug label or professional guidelines) as well as 9 level 3A and 12 level 4 genes (predictive of drug response based on well-powered clinical studies or compelling biological evidence, respectively). OncoKB data is used internally to annotate >15,000 MSK patient sequencing reports annually, and its content is integrated into the cBioPortal for Cancer Genomics. Methods: To quantitate the expansion of the clinical actionability landscape in precision oncology, we annotated the clinical sequencing results of 47,271 solid tumor samples from the AACR Project GENIE cohort (GENIE 11.0 - public) using two versions of OncoKB, the first from March 2017 and the second from November 2022. Results: From 2017 to 2022, we observed a 22.7% increase in the fraction of samples with a standard care biomarker (levels 1 and 2), and a 21.4% decrease in the fraction of samples with a driver alteration that was not clinically actionable. The tumor agnostic approvals of the anti-PD1 antibody pembrolizumab in microsatellite instability high (MSI-H) and tumor mutational burden high (TMB-H) solid tumors resulted in a 11.1%increase of samples with a level 1 biomarker. In sum, of the 38,722 samples across 66 cancer types with limited to no actionability in 2017 (levels 3B, 4 or no level), 21.7% became clinically actionable (levels 1, 2 or 3A) by November 2022. This is an upper bound estimate as it includes samples that may have been eligible for an FDA-approved precision oncology therapy based on cancer subtype alone. Conclusions: Our analysis highlights the significant expansion of precision oncology-based therapeutic options for patients with cancer over the past 5 years, as well as the ongoing unmet need for novel precision oncology approaches designed to address oncogenic but not currently actionable drug targets. [Table: see text]

New ESMO scale for clinical actionability of molecular targets (ESCAT) for gliomas based on a multicentric real world data cohort using next-generation sequencing (NGS).

2055 Background: The ESMO ESCAT scale rank molecular alterations (MA) based on the published evidence supporting their clinical use. Building on our proposal for the first ESCAT classification for primary brain tumors (pBT) [Mirallas et al. ESMO 2022], we aim to describe the clinical actionability of NGS in a multicentric glioma cohort and gather clinical factors that enrich patients with MA to justify performing NGS. Methods: A multicentric retrospective study included pts with molecular profiling using NGS methods (Foundation Medicine, local NGS, Caris, Oncomine, and fusion panels). Clinical actionability was classified using the ESCAT scale; Tier 1 (ready for clinical implementation), Tier 2 (alteration-drug match with antitumor activity), Tier 3 (supported in other tumor types), and Tier 4 (preclinical evidence). Clinical factors considered for enrichment were diagnosis of glioblastoma (GBM), sex, and age ≤40 years, differences between groups were determined using Chi-squared test. The overall survival (OS) was calculated through Kaplan-Meier method and cox hazard ratio were fitted. Results: A total of 361 pts with NGS performed between Feb 2018 and Dec 2022 at 4 hospitals in Spain. Median age was 51.5 (range, 3.3-83.8), 77% had ECOG ≥1, 39.2% were women, 23.4% were 40 or younger, 20.8% had IDH1mut, and 73.5% had a diagnosis of GBM. The distribution of MA according to ESCAT was: 10 pts ESCAT 1 (5 BRAFV600E mut, 5 NTRK 1-3 fusions); 75 pts ESCAT 2 (67 IDH1 mut, 6 FGFR-TACC rearrangement, 4 FGFR 1-3 mut, 3 IDH2 mut); 105 pts ESCAT 3 (98 PIK3CA/PTEN mut, 12 PTEN loss, 11 H3K27M mut, 6 BRAF fusion, 2 MET mut, 1 FGFR3 amplification); 93 pts ESCAT 4 (119 TERT mut, 77 EGFR gain, 67 CDKN2A loss, 58 CDKN2B loss, 48 EGFRvIII rearrangement, 46 ATRX mut, 40 EGFR mut, 8 ATM mut). Prevalence of Tier 1-2 MA was 23.6%. In relation to enrichment factors: 12% >40y pts had ESCAT1-2 vs 61% in ≤40y (p<0.001), pts with GBM had a 11% prevalence vs 61% non-GBM (p<0.001), and non-gender differences were seen. The median OS for Tier 1-2 was 135 months (95% CI 166-99) vs Tier 3-4 of 24 months (95% CI 21.5-29.6). Significantly worse OS was observed for Tier 3-4 gliomas (HR: 4.11; 2.84-5.95, p<0.001), and remained significant after adjusting for histology and hospital center (p<0.01). Conclusions: The prevalence of ESCAT Tier 1-2 alterations is high in gliomas suggesting that molecular profiling should be offered, at least, in tertiary centers where these techniques are available. Potential enrichment factors to offer NGS were age and non-GBM pBT, but not gender.

A next-generation sequencing methylation assay for sub-typing small cell lung cancer in cell-free DNA.

e15021 Background: Liquid biopsy provides an easy access and minimally invasive technique to inform cancer patient management in clinics as well as clinical trials. Detecting genomic alterations in circulating tumor DNA (ctDNA) has demonstrated clinical utility and actionability in therapy selection and longitudinal monitoring. The development of a comprehensive assay that provides multi-omic information is imperative to expand the utility of liquid biopsy. Methods: Agilent Resolution ctDNA diagnostic assays can detect all four major types of actionable genomic alterations for cancer patients: SNVs, Indels, fusions, and CNVs. The Agilent Resolution ctDx FIRST assay has recently been approved by the FDA as a companion diagnostic assay for non-small cell lung cancer (NSCLC). Applying the same core technology, we developed an additional assay that enables methylation profiling in cfDNA. The custom methylation-specific panel targets informative regions to detect tumor signals and potentially profile the disease by subtype. Recent reports suggest that small cell lung cancer (SCLC) may be stratified into subtypes based on gene expression, including that of ASCL-1, NEUROD1, and POU2F3. We used the proposed SCLC subtyping model to test the feasibility of subtyping SCLC by cfDNA methylation profiles. A baseline methylation profile was established using cell-free DNA from healthy donors. Test samples included genomic DNA from SCLC cell lines and plasma samples of SCLC patients with matching tissue, where available. Results: We applied a methylation-specific bioinformatic pipeline that generates error-corrected unique molecules. The methylation workflow showed unique coverage comparable to our ctDx assay and produced consistent results from input DNA as low as 1ng. We found consistent methylation patterns in cfDNA among healthy donors. SCLC cell lines showed distinct profiles that could be classified into sub-types. When spiked into healthy donor cfDNA background, the tumor signals were distinguishable down to ~0.5%. Plasma samples from SCLC patients showed profiles consistent with that of SCLC sub-types defined by cell lines. Matching plasma and tissue samples produced concordant sub-typing results, supporting the feasibility of disease subtyping in both sample types. Samples were also tested by a custom SCLC panel to analyze genomic alterations. Prevalent variants in SCLC were detected, including SNVs in TP53, PIK3CA, RB1, and MYC amplifications. Conclusions: We showed a cfDNA methylation profiling platform that could support disease sub-typing for SCLC. This platform has great potential in disease stratification and patient management.

SO-3 Clinical relevance and actionability of BRAF alterations in advanced biliary tract cancer: Preliminary results from the multicenter B-REAL study

SO-3 Clinical relevance and actionability of BRAF alterations in advanced biliary tract cancer: Preliminary results from the multicenter B-REAL study

Korean Precision Medicine Networking Group study of molecular profiling guided therapy based on genomic alterations in advanced solid tumors (KOSMOS) II (KCSG AL 22-09).

TPS1608 Background: Next-generation sequencing (NGS) has been introduced to many Korean institutions as molecular diagnostics for cancers since 2017, when it was reimbursed by the National Health Insurance Service. However, molecularly-guided treatment (MGT) options based on NGS results have been limited due to the strict regulation for prescribing outside of the approved indication, lack of clinical trial availability, and limited access to molecular tumor boards (MTB) in most institutions. Given our previous KOSMOS study results, which showed MGT was available for 54% (124/198) of patients based on local NGS (ESMO Open 2022;7(6, Suppl 1);100653), KOSMOS-II study was designed to demonstrate MGT would be feasible and effective via a nation-wide precision medicine platform. Methods: This national, prospective, pragmatic, multi-cohort study comprised of a framework to screen pts with metastatic solid tumors for actionable genetic alterations (GA) based on local NGS testing and to recommend MGT by a remote and centralized MTB held weekly or biweekly. MGT can be one of the following options: Tier 1, the therapeutic use of investigational drugs targeting GAs including ALK, EGFR, ERBB2, BRAF, FH, ROS1, RET, or high tumor mutational burden; Tier 2, drugs that are allowed for treatment outside of the approved indications by the National Health Insurance; Tier3, clinical trials matched for GA recommended by MTB. To estimate the proportion of patients receiving MGT to be 50±3.25% with 95% confidence interval, we will enroll 1,000 pts, considering 5% drop-out rate. The clinical benefit rate of MGT beyond 16±2 weeks is the second co-primary endpoint, which is expected to be 30% with type I error of 5% and 90% power. Pts must have progressed ≥ 1 or more lines of therapies; have undergone NGS testing; are able to provide the Variant Call Format (vcf) file of the NGS results. MTB determines clinical actionability of GA and recommends MGT if possible, according to Tier 1,2, or 3. The vcf files and matched clinical information will be deposited to the repository of National Center Center Korea and will be curated in a de-identified clinico-genomic database. As of Feb 2023, 139 of planned 1,000 have been enrolled. Clinical trial information: NCT05525858 .

Clinical and genomic characterization of early-onset pancreatic cancer.

4018 Background: Pancreatic cancer is a leading cause of cancer-related death worldwide. Early-onset tumours (≤ 50 years) have risen alarmingly in recent years. However, the clinical and genomic particularities of early-onset pancreatic cancer (EOPC) remain poorly defined. We aimed to characterize EOPC and study its implications for treatment and prognosis. Methods: We performed a retrospective analysis of EOPC patients and a control group of patients ≥ 70 years (defined as average-onset pancreatic cancer, AOPC) followed at a tertiary cancer centre from 2010 to 2022. We collected baseline patient characteristics, tumor molecular profiling, germline genetic alterations, survival and treatment outcomes. We used a targeted gene panel to identify somatic genomic events and classified them according to the ESMO scale for clinical actionability of molecular targets (ESCAT). Key molecular findings were validated in an external cohort. We used a propensity score weighting method and multivariate Cox regression analysis to adjust for covariates. Results: We reviewed 824 patients, 336 of whom met all inclusion criteria (EOPC N = 139, AOPC N = 197). EOPC was associated with smoking status (current, 15.9 vs 7.9%, p = 0.03), lower prevalence of diabetes (3.7 vs 39.6%, p < 0.01), better performance status (ECOG 0, 42.9 vs 19.3%, p < 0.01), higher CA19.9 levels (median 574 vs 207.6UI/L, p = 0.07) and higher albumin levels (median 4.2 vs 3.9g/L, p < 0.01). EOPC showed a non-significant higher prevalence of germline alterations (22.4 vs 14.5%, p = 0.18). After adjustment for baseline covariates, we observed no differences in survival (HR 0.94, 95% CI 0.63-1.4). We found a lower prevalence of KRASMUT tumors in EOPC when compared with AOPC (83.1 vs 91.1%, p = 0.12) and validated this finding in an independent cohort of 803 patients (82.9 vs 94.5%, p < 0.01). Notably, EOPC were enriched in potentially actionable alterations when compared with AOPC, both in our cohort (19.1 vs 14.4%) and the validation cohort (14.4 vs 7.9%, p < 0.01). Moreover, 294 patients in our cohort (EOPC N = 130; AOPC N = 164) were diagnosed with or eventually presented metastasis. EOPC more frequently received first-line 5FU-based chemotherapy (44.1 vs 14.7%, p < 0.01) and received a greater number of treatment lines (median 2 vs 1, p < 0.01). EOPC had a longer progression-free survival on first-line chemotherapy when compared with AOPC (HR 0.61, 95% CI 0.43-0.87, p < 0.01) although there were no differences in response rate. Six EOPC patients received matched targeted therapies and 5 patients with AOPC. After adjusting for the number of treatment lines received, EOPC treated with targeted therapies exhibited longer OS compared with EOPC who did not (HR 0.34 95% CI 0.12-0.93, p = 0.04), whereas this trend was not observed in AOPC (HR 0.82 95% CI 0.32-2.11, p = 0.68). Conclusions: EOPC patients harbor unique clinical and molecular features and may particularly benefit from precision-based oncology approaches.

Concordance of circulating and solid tumor DNA through comprehensive genomic profiling in a large integrated cancer network.

3059 Background: Comprehensive Genomic Profiling (CGP) of solid tumors is crucial in diagnosing and treating cancer patients. Circulating Tumor DNA (ctDNA) is a potential biomarker for detecting mutations when tissue samples are limited or unavailable. However, the sensitivity of ctDNA tests vary greatly, making concordance studies with tissue samples critical to validate clinical use. This study presents results of CGP in paired tumor tissue and plasma samples from representative cancer patients treated in a community-based, Integrated Network Cancer Program. Methods: Over 2000 cancer patients have provided solid tumor and/or blood samples to date including serial follow-up blood samples. To optimize ctDNA integrity, we collected blood using a membrane stabilizer (Streck), ensured courier transport in ≤ 72 hours, separated plasma via differential centrifugation x 3, extracted cell free DNA with magnetic beads, and performed Next Generation Sequencing (NGS) using the ct-TSO500 assay (Illumina). In cases where patients provided both a solid tumor and concurrent blood sample, st-TSO500 NGS results from the solid tumor were compared to the ctDNA results for these matching specimens. Clinical actionability of variants was determined using the OncoKB database at levels 1, 2, and R1. Results: 146 matched blood and solid tumor specimens were analyzed to date. CGP studies revealed 1179 ± 102 unique variants across 20 different cancer types. Coding mutations from these specimens averaged 183 ± 25 variants. The ctDNA contained 96.5% of initial variants (1088 ± 95) and 95.1% of coding mutations (163 ± 20) found in the respective matched solid tumor. 76 patients had one or more oncogenic mutations detected only in the ctDNA and 28 (36.8%) of these included an actionable anticancer pharmacotherapy. Concordance by selective tumor type for clinical oncogenic variants was 95.8% (prostate), 95.0% (pancreas), 92.8% (lung and ovarian), 90.0% (endometrium), 84.3% (colon) and 82.4% for breast cancer. There was no difference in concordance between stage IV (91.0%) and stage I-III (86.5%) tumors (p = 0.14). An average of 6.2 ± 2.5 oncogenic variants were detected in both ctDNA and tissue. 99.3% of all patients contained an oncogenic biomarker in both ctDNA and tissue assays. Conclusions: We developed a prospective tissue and plasma repository of patients across a large integrated cancer network. These data demonstrate high concordance between circulating and solid tumor DNA using an “in-house” CGP assay. With no loss of concordance across cancer stages, ctDNA can identify actionable mutations not found in solid tumor analysis. This study is an important step toward incorporating “in-house” ctDNA CGP into routine cancer care. Future goals are to utilize these data for development of novel prognostic and predictive classifiers for use in a community-based hospital system across various disease sites.

The mutational landscape of circulating cell free DNA in patients with gastric cancer.

e16055 Background: Gastric cancer (GC) is a malignant tumor with a high lethality rate, with the highest incidence occurring in China. Gastric cancer is phenotypically heterogeneous, therefore the identification of genomic alteration is crucial for guiding clinical treatment. In this study, we aimed to use liquid biopsies to detect somatic mutations of GC patients and describe the mutational landscape of circulating tumor DNA. Methods: The blood samples of 157 Chinese patients with gastric malignancies were collected between 2018 to 2022. Somatic alterations were detected by either whole exome sequencing or high-depth targeted gene panel sequencing of ctDNA, using white blood cells as germline DNA control. Results: In our cohort, 1002 alterations of 323 genes were detected from plasma samples. We used MutSigCV tool to define the significantly mutated genes (SMG) and identified 16 SMG. Overall, the five most common mutated genes were TP53 (52.6%), LRP1B (15.4%), ARID1A(14.7%), SYNE1(12.2%), CDH1(10.9%). Recurrent mutations were identified in TP53 genes including c.524G > A, c.713G > T, which lead to loss of function of P53 protein according to evidence of clinical actionability from OncoKB (http://oncokb.org/). Genetic amplification of ERBB2 was detected in 8 patients (5.1%). 6 out of 95 samples were microsatellite instability (MSI), 3 were MSI-H and 3 were MSI-L.The median blood tumor mutation burden (bTMB) of all samples was 1.97 (range 0–143.51) mutations/Mb, 2 patients (1.27%) to high bTMB (≥ 20 mutations/Mb), 37 patients (23.57%) to intermediate bTMB (5-19 mutations/Mb), 118 patients (75.16%) to low bTMB ( < 5 mutations/Mb). The Co-occurrence and mutual exclusivity of gastric cancer variations was investigated. Mutually exclusive mutation patterns were observed in TP53-KRAS (p < 0.05), TP53-ARID1A (p < 0.05), TP53-CDH1 (p < 0.05) gene pairs. In addition, Gene Ontology (GO) and KEGG analysis were performed to analyze enrichment of somatic mutated genes and understand their biological function. The altered signaling pathway included PI3K−Akt signaling, MAPK signaling, Ras signaling, Rap1 signaling etc. 61 genes with somatic mutations were enriched to PI3K–Akt signaling pathway. Conclusions: In this study, we characterized the somatic mutational landscape of gastric cancer in 157 Chinese patients utilizing ctDNA which might guide clinical treatment.